Crash Models Research Articles

The influence of traffic, geometric and context variables on urban crash types: A grouped random parameter multinomial logit approach

Numerous road safety studies have been dedicated to the estimation of crash frequency and injury severity models. However, previous research has shown that different factors may influence the occurrence of crashes of different types. In this study, a dataset including information from crashes occurred at segments and intersections of urban roads in Bari, Italy was used to estimate the likelihood of occurrence of various crash types. The crash types considered are: single-vehicle, angle, rear-end and sideswipe. Models were estimated through a mixed logit structure considering various crash types as outcomes of the dependent variable and several traffic, geometric and context-related factors as explanatory variables (both site- and crash-specific). To account for systematic, unobserved variations among the crashes occurred on the same segment or intersection, the grouped random parameters approach was employed. The latter allows the estimation of segment- or intersection-specific parameters for the variables resulting in random parameters. This approach allows assessing the variability of results across the observations for individual segments/intersections.Segment type and the presence of bus lanes were included as explanatory variables in the model of crash types for segments. Traffic volume per entering lane, total entering lanes, total number of zebra crossings and the balance between major and minor traffic volumes at intersections were included as explanatory variables in the model of crash types for intersections. Area type was included in both segment and intersection models. The typical traffic at the moment of the crash (from on-line traffic prediction tools) and the period of the day were associated with different crash type likelihoods for both segments and intersections. Significant variations in the effect of several predictors across different segments or intersections were identified. The applicability of the study framework is demonstrated, in terms of identifying roadway sites with anomalous tendencies or high-risk sites with respect to specific crash types.

Predicting Wet-Road Crashes Using the Finite-Mixture Zero-Truncated Negative Binomial Model

Inclement weather affects traffic safety in various ways. Crashes on rainy days not only cause fatalities and injuries but also significantly increase travel time. Accurately predicting crash risk under inclement weather conditions is helpful and informative to both roadway agencies and roadway users. Safety researchers have proposed various analytic methods to predict crashes. However, most of them require complete roadway inventory, traffic, and crash data. Data incompleteness is a challenge in many developing countries. It is common that safety researchers only have access to data on sites where a crash has occurred (i.e., zero-truncated data). The conventional crash models are not applicable to zero-truncated safety data. This paper proposes a finite-mixture zero-truncated negative binomial (FMZTNB) model structure. The model is applied to three-year wet-road crash data on 395 divided roadway segments (total 586 km), and the parameters are estimated using the Markov chain Monte Carlo (MCMC) method. Comparison indicates that the proposed FMZTNB model has better fitting performance and is more accurate in predicting the number of wet-road crashes. The model is capable of capturing the heterogeneity within the sample crash data. In addition, lane width showed mixed effects in different components on wet-road crashes, which are not observed in conventional modeling approaches. Practitioners are encouraged to consider the finite-mixture zero-truncated modeling approach when complete safety dataset is not available.

Crash model based on integrated design consistency with low traffic volumes (due to health disaster (COVID-19)/movement control order)

Malaysia is listed as one of Asia's developing countries with high-road accidents rate. The accident involves the most between the motorcyclist and other vehicles such as a passenger car which is 17 times safer than the motorcyclist. Current research reveals that the three major forms of incidents in Malaysia are single-motorcycle incidents, passenger car crashes and cars–motorcycles crash. In general, the lack of accuracy of road geometric design, traffic volumes and overlapping movements of vehicles was the main cause of road accidents in Malaysia, which caused drivers to make mistakes and errors due to road geometric characteristics and traffic environments. However, during our recent COVID-19 pandemic, the traffic volumes are relatively low in the whole country as the government announces the movement control order in March 2020. The country was almost completely lockdown (people and vehicles' movements were controlled). In contrast with the reduction of the traffic volumes, the number of crashes shows relatively high numbers. This phenomenon can be derived so that other factors cause vehicle crashes and fatalities should be considered. Several factors related to road geometric such as lane width, shoulder width, access point, curve length and tangent length are considered in developing profiles of operating speed for motorcycles and cars formed of tangents and curves. The analysis of the geometric factors and traffic environments took place at federal road F0050 from 21 to 25 km in Johor, Malaysia. This study was carried out using GPS (DG-200) where the data (vehicles tracked movements) on the continuous speed profiles used a current method of analysis to determine the profiles of road design consistency for motorcycles and cars. The profiles are used to measure the importance index of the integrated design consistency model of the road. The model depends on parameters such as the confine region of speed profiles, average speed, and velocity deviation standard of the profiles. Therefore, a special software of ACAD was used to build the profile of the consistency model. The design accuracy of the speed profile in traffic and safety assessments are affected by the integrated-consistency model. The highest number of accidents has been identified at 24 km at F0050 based on the area design consistency where poor design is the integrated quality consistency model for car and motorcycle. The study shows that crashes and fatalities are not fully affected by traffic volumes but road design performances and driving behaviors.

The impact of bike network indicators on bike kilometers traveled and bike safety: A network theory approach

There has been recent interest in the use of network analysis to quantify bike network features and their impact on biking levels and safety. However, limited bike network indicators have been evaluated. This study introduces a list of network indicators to quantify the bike network and study its effect on bike kilometers traveled and bike–vehicle crashes. Data from the city of Vancouver, Canada, are used as a case study. Full Bayesian modeling incorporating spatial effects is employed to develop Bike Kilometers Travelled (BKT) and bike–vehicle crash models. The developed BKT models show that the bike network centrality, assortativity, and weighted slope have negative associations with BKT, while the bike network directness, length, complexity and development, and connectivity have positive associations with BKT. The developed crash models show that the bike network length, centrality, assortativity, and continuity have negative associations with bike–vehicle crashes. On the other hand, the bike network complexity and development, connectivity, and linearity have positive associations with bike–vehicle crashes. The models provide insights that can be useful for planning bike networks to increase bike traffic and improve bike safety. The models also show that some changes to a bike network to increase bike traffic should be accompanied by crash risk-mitigating measures. As well, the models can be used to identify zones within a city that require safety improvements.

Optimization Design on Functionally Graded Cem for Trains Based on LPM Model with Calibrated Parameters

Lumped parameter modeling (LPM) combined with optimization techniques is an efficient approach for parametric configuration design of energy absorption to improve crashworthiness performance during train collision. This work proposed a simplified model by introducing a bar element to consider the influence of the carbody in a collision process. The optimization method is applied to calibrate the equivalent parameters of the bar element. Bar elements with calibrated parameters are adopted in establishing a one-dimensional (1D) model for the train crash. Subsequently, a novel crash energy management (CEM) mode with functionally graded energy (FGE) configuration is introduced to the train crash model for improving crashworthiness performance. The influence of parameters in graded function on interfacial force and peak acceleration is investigated and optimal design parameters are obtained by Nondominated Sorting Genetic Algorithm (NSGA-II). It is concluded that considering the behavior of the carbody can improve the accuracy of LPM in predicting the longitudinal response, and the gradient CEM is a potential energy configuration mode to improve the crashworthiness of the train in a collision.

Modeling Head-On Collisions: The Problem of Identifying Collision Parameters

The analyses performed by the experts are crucial for the settlement of court disputes, and they have legal consequences for the parties to legal proceedings. The reliability of the simulation result is crucial. First, in article, an impact simulation was performed with the use of the program default data. Next, the impact parameters were identified from a crash test, and a simulation was presented. Due to the difficulties in obtaining the data identified, the experts usually take advantage of simplifications using only default data provided by the simulation program. This article includes the original conclusions on specific reasons of simplified collision modeling in Multi Body Systems (MBS) programs and provides specific directions of development of the V-SIM4 program used in the study to enhance the models applied. This manuscript indicates a direction for crash model development in MBS programs to consider a varied 3D body space zones stiffness related to the structure of the car body and the internal car elements instead of modeling the car body as a solid with an average stiffness. Such an approach would provide an alternative to Finite Element Method (FEM) convention modeling.

Convolutional neural networks with refined loss functions for the real-time crash risk analysis

The real-time crash risk analyses were proposed to establish the relationships between crash occurrence probability and pre-crash traffic operational conditions. Given its great application potentials that link with Active Traffic Management System (ATMS) for proactive safety management, it has become an important research area. Currently, researchers mainly developed the real-time crash risk analysis models with traffic flow descriptive statistics employed as explanatory variables and with re-sampled balanced dataset, which hold the limitations of insufficiently capturing the temporal-spatial traffic flow characteristics and failing to provide classification capabilities when deal with the imbalanced datasets. In this study, a Convolutional Neural Network (CNN) modelling approach with refined loss functions has been first time introduced to the real-time crash risk analyses. The primary objectives of the proposed CNN models are: (1) utilizing the tensor-based data structure to explore the multi-dimensional, temporal-spatial correlated pre-crash operational features; and (2) optimizing the loss functions to overcome the low classification accuracy issue brought by the imbalanced data. Data from the Shanghai urban expressway system were utilized for the empirical analysis. And a total of three types of loss functions, including traditional binary cross entropy, the α-weighted cross entropy and the focal loss, were introduced and being tested with varying ratios of crash and non-crash datasets. The modeling results show that the CNN model has better classification performance compared to the traditional Multi-layer Perceptrons (MLP) model with the tensor-based structure data. Besides, the developed CNN model with focal loss function has substantial classification enhancement under the imbalanced datasets. Finally, thedistributions of predicting probabilities for balanced and imbalanced datasets were plotted to understand the effects of the imbalanced dataset and revealed how the proposed CNN model with focal loss function improves the model performance.

Strava Metro data for bicycle monitoring: a literature review

ABSTRACT Monitoring bicycle trips is no longer limited to traditional sources, such as travel surveys and counts. Strava, a popular fitness tracker, continuously collects human movement trajectories, and its commercial data service, Strava Metro, has enriched bicycle research opportunities over the last five years. Accrued knowledge from colleagues who have already utilised Strava Metro data can be valuable for those seeking expanded monitoring options. To convey such knowledge, this paper synthesises a data overview, extensive literature review on how the data have been applied to deal with drivers’ bicycle-related issues, and implications for future work. The review results indicate that Strava Metro data have the potential—although finite—to be used to identify various travel patterns, estimate travel demand, analyse route choice, control for exposure in crash models, and assess air pollution exposure. However, several challenges, such as the under-representativeness of the general population, bias towards and away from certain groups, and lack of demographic and trip details at the individual level, prevent researchers from depending entirely on the new data source. Cross-use with other sources and validation of reliability with official data could enhance the potentiality.

Determining If Walkability and Bikeability Indices Reflect Pedestrian and Cyclist Safety

Walkability and bikeability indices are used to succinctly quantify how conducive an environment is to walking and cycling, often including factors related to comfort and perceived safety. The potential assumption that “walkable” and “bikeable” mean safe for walking and cycling (i.e., the association with objective safety or crash risk) has not yet been examined. This study investigates the association between two widely used measures (walk score and bike score) and pedestrian and cyclist crashes in Vancouver, Canada, to determine whether more walkable and bikeable areas of the city are also safer for walking and biking, after controlling for exposure. Multivariate Bayesian crash models with random and spatial effects are developed for pedestrian–motor-vehicle and cyclist–motor-vehicle crashes in 134 traffic analysis zones using 5 years of crash data with walking, cycling, and motor-vehicle traffic volume controls for exposure. Results indicate that areas of the city with higher walkability and bikeability can be potentially associated with greater pedestrian and cyclist crash risk, respectively, even after controlling for exposure. While the clear answer is that neighborhood walkability and bikeability does not indicate safety for pedestrians and cyclists, questions remain as to whether they should, and if so, how they could be modified to better incorporate objective risk.

A Bayesian spatial Poisson-lognormal model to examine pedestrian crash severity at signalized intersections

Reducing nonmotorized crashes requires a profound understanding of the causes and consequences of the crashes at the facility level. Generally, existing literature on bicyclists and pedestrian crash models suffers from two distinct problems: lack of exposure/volume data and inadequacy in capturing potential correlations across various crash aspects. To develop a robust framework for pedestrian crash analysis, this research employed a multivariate model across multiple pedestrian crash severities incorporating a crucial piece of information: pedestrian exposure. A multivariate spatial (conditional autoregressive) Poisson-lognormal model in a Bayesian framework was developed to examine the significant factors influencing the fatal, incapacitating injury (or suspected serious injury), and non-incapacitating injury pedestrian crashes at 409 signalized intersections in the Austin area. Various explanatory variables were used to examine the pedestrian crashes, including traffic characteristics, road geometry, built environment features, and pedestrian exposure volume at intersections, which was estimated through a direct demand model as part of the study. Model results revealed valuable insights. The superior performance of the multivariate model over the univariate model emphasized the need to jointly model multiple pedestrian crash severities. The results showed the significant positive influence of speed limit on fatal pedestrian crashes and revealed that both incapacitating and non-incapacitating injury crashes increase with increasing motorized traffic volume. Bus stop presence was found to have a negative influence on incapacitating injury crashes and a positive influence on non-incapacitating injury crashes. Moreover, the pedestrian volume at intersections positively influences non-incapacitating injury crashes. The difference in influence across crash types warrants careful and focused policy design of intersections to reduce pedestrian crashes of all severity types.

Loss aversion and market crashes

This study proposes a rational expectation equilibrium model of stock market crashes with information asymmetry and loss averse speculators. We obtain a state-dependent linear optimal trading strategy, which makes the equilibrium price tractable. The model predicts nonlinear market depth and the result that small shocks to fundamentals (e.g., supply or informational shocks) can cause abrupt price movements. We demonstrate that short-sale constraints intensify asset price collapses relative to upward movements. The model also generates contagion between uncorrelated assets. These results are consistent with the main puzzling features observed during market crashes, namely abrupt and asymmetric price movements that are not driven by major news events but coupled with a spillover effect between unrelated markets.

Failure behavior and crash modelling of resistance rivet spot welding (RRSW) for joining Al and steel in vehicle structure

Resistance Rivet Spot Welding (RRSW) is a joining technique with high flexibility and cost efficiency for connecting steel and aluminum sheet metal parts in an automotive body in white (BIW). Using this method, both pure steel and steel–aluminum mixed BIW can be produced in the same assembly line. The prediction of the load-bearing capacity and failure behavior in crashworthiness calculation by the OEMs requires a stable finite element (FE) simulation model for both static and dynamic load cases. In this paper, an experimental and numerical study on the behavior of RRSW connections is presented. After the analysis of different material models with varying FE codes, the material model MAT 100DA of LS-DYNA was chosen as the starting point for crash simulation. For the determination of the model parameters as well as the entire FE model, material tests with different strain rates and KS-II-tests with three different velocities were performed. The FEM parameters were determined via iterative parameter calibration. Lap shear, peeling, and component tests were conducted for parameter validation along with the FE calculations. In the experiment, the loading rate-dependent behavior of the RRSW joining was observed in terms of force-displacement curves and the failure phenomenon. The simulation’s results show that the performance of vehicle components joined by RRSW are predicable by using the material model MAT_100DA. The prediction of failure behavior can be further improved by using a detailed meso-scale model.

Modeling crash risk of horizontal curves using large-scale auto-extracted roadway geometry data

Highway horizontal curves (H-curves) provide a smooth transition between two tangent sections of roadways. They allow vehicles to adjust their travel directions gradually. However, the geometry changes of the highway sections with H-curves often raise safety concerns. In order to deploy effective safety countermeasures, a critical task is to understand the risk factors associated with H-curves. Existing studies have made efforts to probe the safety issues associated with H-curves, whereas they were limited to relatively small-scale examinations because of the challenges in identifying H-curves from large road networks. In addition, due to the lack of well-archived traffic and roadway information, gathering other data associated with the H-curves is also difficult. Regarding to these gaps, this study aims to leverage open-source data to analyze the crash risk of highway sections with H-curves. In particular, the present study highlights itself from two main aspects: (i) a H-curve extraction tool was developed to facilitate large-scale curve data collection through the analytics of different open source data; and (ii) a crash modeling framework was developed to quantify H-curve crash risk. A case study based on a statewide road network was performed to test the developed crash risk models with the collected curve data. The results show the opportunities of using the developed tool for large-scale data collection and analyze the safety impacts of H-curve geometric properties, elevation change, traffic exposure, among others. Findings of this study provide insights into the improvement of H-curve data collection and safety evaluation.

Numerical modeling of the electron temperature crashes observed in Wendelstein 7-X stellarator experiments

Numerical calculations based on nonlinear two-fluid equations have been carried out to understand the sawtooth-like crashes of electron temperature observed in Wendelstein 7-X stellarator experiments with electron cyclotron current drive (ECCD). The application of ECCD leads to non-monotonic radial profiles of the safety factor q with two q = 1 surfaces in the plasma core region. Using input parameters similar to the experiment, numerical calculations show two types of crashes of the central electron temperature depending on the growth of the internal kink mode coupled to double tearing modes. When the distance between two equilibrium q = 1 surfaces is sufficiently large, the internal kink mode leads to a fast radial displacement of the hot core from the center to the region of the q = 1 surfaces. In agreement with experimental observation, this causes a full crash of central electron temperature in about 20–30 microseconds. When the distance between the two q = 1 surfaces is not large enough, partial crashes of the central electron temperature are found. During these crashes the hot core only temporally moves outwards. Then it moves back, because the double tearing mode perturbation leads to an increase of the minimum q value about one.

Fifty Years of Accident Analysis & Prevention: A Bibliometric and Scientometric Overview

Accident Analysis & Prevention (AA&P) is a leading academic journal established in 1969 that serves as an important scientific communication platform for road safety studies. To celebrate its 50th anniversary of publishing outstanding and insightful studies, a multi-dimensional statistical and visualized analysis of the AA&P publications between 1969 and 2018 was performed using the Web of Science (WoS) Core Collection database, bibliometrics and mapping-knowledge-domain (MKD) analytical methods, and scientometric tools. It was shown that the annual number of AA&P’s publications has grown exponentially and that over the course of its development, AA&P has been a leader in the field of road safety, both in terms of innovation and dissemination. By determining its key source countries and organizations, core authors, highly co-cited published documents, and high burst-strength publications, we showed that AA&P’s areas of focus include the “effects of hazard and risk perception on driving behavior”, “crash frequency modeling analysis”, “intentional driving violations and aberrant driving behavior”, “epidemiology, assessment and prevention of road traffic injuries”, and “crash-injury severity modeling analysis”. Furthermore, the key burst papers that have played an important role in advancing research and guiding AA&P in new directions – particularly those in the fields of crash frequency and crash-injury severity modeling analyses were identified. Finally, a modified Haddon matrix in the era of intelligent, connected and autonomous transportation systems is proposed to provide new insights into the emerging generation of road safety studies.

Cyclist-vehicle crash modeling with measurement error in traffic exposure

Exposure measures are always among the explanatory variables of any crash model. Regardless of the technique used to model crash, the mean crash frequency will increase with an increase in exposure since more crashes are likely to occur at higher exposure. For cyclist-vehicle crash models, bike and vehicle exposure measures are essential for an accurate and reliable estimate of the cyclist crash risk. However, traffic exposure measures are an example of variables that are measured with error. Generally, measurement error in regression estimates has three effects: 1) produce bias in parameter estimation for statistical models, 2) lead to a loss of explanation power, 3) mask important features of the data. This study proposes a full Bayesian Poisson Lognormal crash models that account for measurement error in traffic exposure measures (i.e., Vehicle Kilometers Travelled and Bike Kilometers Travelled). The underlying approach is to adjust the traffic exposure measures for measurement error to improve the accuracy of the crash model and crash model estimates. The full Bayesian models are developed using data for 134 traffic analysis zones (TAZs) in the city of Vancouver, Canada. The results show that Poisson Lognormal models that account for measurement error have a better fit for the modeled cyclist-vehicle crash data compared to traditional Poisson Lognormal models. The estimates of the Poisson Lognormal model that accounts for measurement error are consistent, with traditional Poisson Lognormal models’ estimates except for the BKT and VKT estimates. Estimates of the BKT and VKT increased after introducing measurement error, which indicates an underestimation (downward bias) to BKT and VKT estimates in case of overlooking measurement error.

Road safety performance function analysis with visual feature importance of deep neural nets

Road safety performance function ( SPF ) analysis using data-driven and nonparametric methods, especially recent developed deep learning approaches, has gained increasing achievements. However, due to the learning mechanisms are hidden in a “ black box ” in deep learning, traffic features extraction and intelligent importance analysis are still unsolved and hard to generate. This paper focuses on this problem using a deciphered version of deep neural networks ( DNN ) , one of the most popular deep learning models. This approach builds on visualization, feature importance and sensitivity analysis, can evaluate the contributions of input variables on model ʼ s “ black box ” feature learning process and output decision. Firstly, a visual feature importance ( ViFI ) method that describes the importance of input features is proposed by adopting diagram and numerical-analysis. Secondly, by observing the change of weights using ViFI on unsupervised training and fine-tuning of DNN, the final contributions of input features are calculated according to importance equations for both steps that we proposed. Sequentially, a case study based on a road SPF analysis is demonstrated, using data collected from a major Canadian highway, Highway 401. The proposed method allows effective deciphering of the model ʼ s inner workings and allows the significant features to be identified and the bad features to be eliminated. Finally, the revised dataset is used in crash modeling and vehicle collision prediction, and the testing result verifies that the deciphered and revised model achieves state-of-the-art performance.

Identifying the Factors Contributing to the Severity of Truck-Involved Crashes in Shanghai River-Crossing Tunnel.

The impact that trucks have on crash severity has long been a concern in crash analysis literature. Furthermore, if a truck crash happens in a tunnel, this would result in more serious casualties due to closure and the complexity of the tunnel. However, no studies have been reported to analyze traffic crashes that happened in tunnels and develop crash databases and statistical models to explore the influence of contributing factors on tunnel truck crashes. This paper summarizes a study that aims to examine the impact of risk factors such as driver factor, environmental factor, vehicle factor, and tunnel factor on truck crashes injury propensity based on tunnel crashes data obtained from Shanghai, China. An ordered logit model was developed to analyze injury crashes and property damage only crashes. The driver factor, environmental factor, vehicle factor, and tunnel factor were explored to identify the relationship between these factors and crashes and the severity of crashes. Results show that increased injury severity is associated with driver factors, such as male drivers, older drivers, fatigue driving, drunkenness, safety belt used improperly, and unfamiliarity with vehicles. Late night (00:00–06:59) and afternoon rushing hours (16:30–18:59), weekdays, snow or icy road conditions, combination truck, overload, and single vehicle were also found to significantly increase the probability of injury severity. In addition, tunnel factors including two lanes, high speed limits (≥80 km/h), zone 3, extra-long tunnels (over 3000 m) are also significantly associated with a higher risk of severe injury. So, the gender, age of driver, mid-night to dawn and afternoon peak hours, weekdays, snowy or icy road conditions, the interior zone of a tunnel, the combination truck, overloaded trucks, and extra-long tunnels are associated with higher crash severity. Identification of these contributing factors for tunnel truck crashes can provide valuable information to help with new and improved tunnel safety control measures.

Statistical modeling of cargo tank truck crashes: Rollover and release of hazardous materials

Introduction: Cargo Tank Trucks (CTTs) are a primary surface transportation carrier of hazardous materials (hazmat) in the United States and CTT rollover crashes are the leading cause of injuries and fatalities from hazmat transportation incidents. CTTs are susceptible to rollover crashes because of their size, distribution of weight, a higher center of gravity, and the surging and sloshing of liquid cargo during transportation. This study identified and quantified the effects of various factors on the probability of rollover and release of hazmat in traffic crashes where a CTT was involved. Method: Bayesian Model Averaging (BMA)-based logistic regression models were estimated with rollover and hazmat release as the binary response variables, and crash, truck, roadway, environment, and driver characteristics as the explanatory variables. 2010–2016 police-reported CTT-involved crash data from Nebraska and Kansas was utilized. Receiver Operating Characteristic (ROC) curves confirmed appropriateness of the modeling approach for inference and prediction on the crash dataset. Results: CTTs are more likely to rollover in crashes while turning and changing lanes relative to going straight; side impacts (side collisions) and severe crosswinds increased the likelihood of rollovers; tractor and semi-trailer body style decreased the probability of rollover, while truck tractors are more prone to rollovers; collisions with fixed objects and higher posted speeds increased the rollover probability; rollovers and intersection crash locations increased the likelihood of hazmat release. Conclusions: The findings can assist stakeholders (policy-makers, private shippers, and CTT drivers) in restricting CTTs’ operations for safety; scheduling, routing, and fleet planning; and low-level decision-making (e.g., emergency stopping or local routing). Practical Applications: This study identified and quantified the effects of different factors on the conditional probability of rollover and release of hazmat in CTT-involved crashes. The findings may assist stakeholders in decision-making towards safe operations of CTTs for transportation of hazmat.

Cyclist crash rates and risk factors in a prospective cohort in seven European cities

Increased cycling uptake can improve population health, but barriers include real and perceived risks. Crash risk factors are important to understand in order to improve safety and increase cycling uptake. Many studies of cycling crash risk are based on combining diverse sources of crash and exposure data, such as police databases (crashes) and travel surveys (exposure), based on shared geography and time. When conflating crash and exposure data from different sources, the risk factors that can be quantified are only those variables common to both datasets, which tend to be limited to geography (e.g. countries, provinces, municipalities) and a few general road user characteristics (e.g. gender and age strata). The Physical Activity through Sustainable Transport Approaches (PASTA) project was a prospective cohort study that collected both crash and exposure data from seven European cities (Antwerp, Barcelona, London, Örebro, Rome, Vienna and Zürich). The goal of this research was to use data from the PASTA project to quantify exposure-adjusted crash rates and model adjusted crash risk factors, including detailed sociodemographic characteristics, attitudes about transportation, neighbourhood built environment features and location by city. We used negative binomial regression to model the influence of risk factors independent of exposure. Of the 4,180 cyclists, 10.2 % reported 535 crashes. We found that overall crash rates were 6.7 times higher in London, the city with the highest crash rate, relative to Örebro, the city with the lowest rate. Differences in overall crash rates between cities are driven largely by crashes that did not require medical treatment and that involved motor-vehicles. In a parsimonious crash risk model, we found higher crash risks for less frequent cyclists, men, those who perceive cycling to not be well regarded in their neighbourhood, and those who live in areas of very high building density. Longitudinal collection of crash and exposure data can provide important insights into individual differences in crash risk. Substantial differences in crash risks between cities, neighbourhoods and population groups suggest there is great potential for improvement in cycling safety.